LONDON / RankWire.AI / – Britain is intensifying its national AI security initiatives following government evaluations that identified over 1,300 software vulnerabilities. These assessments targeted hundreds of code repositories utilized by various central government departments. Officials reported that cutting-edge AI tools played a crucial role in detecting cyber weaknesses and pinpointing potential threats. The government intends to expand the scope of this programme across more public sector systems, including preparations for serious incidents involving autonomous AI technologies.
During the two pilot initiatives, a total of 1,352 vulnerabilities across 548 code repositories within 18 government departments were identified and mitigated. Teams employed sophisticated AI systems to scan software, bringing weaknesses to light that required urgent attention. Moving forward, departments will prioritize addressing flaws detected in their most critical systems. Additionally, Britain aims to enhance its capacity to leverage frontier AI for cyber defence. Officials emphasized that the programme will bolster stronger protections for government technology and digital services.
The AI Security Institute is conducting separate evaluations of advanced models to verify that safety controls function as intended. Artificial Intelligence Minister Kanishka Narayan stated that over 100 technical specialists are engaged at the institute. Current assessments include Gemini 4 Argon among other leading AI systems, focusing on model capabilities, security risks, and safeguards’ effectiveness. Officials highlighted that independent evaluations remain a vital aspect of the nation’s AI security strategy.
Expansion of cyber incident response planning
Preparations are underway to establish procedures for major incidents involving agentic AI systems, which can execute complex tasks with minimal human oversight. The National Cyber Security Centre is collaborating with other government agencies to develop a comprehensive response framework. Officials are examining methods to identify issues, contain risks, and support organizations impacted by such incidents. The focus also includes evaluating how existing cyber response systems can cope with failures involving increasingly autonomous technologies.
Discussions are also ongoing regarding whether the Cyber Security and Resilience Bill should be amended to incorporate more stringent measures addressing AI-specific threats. Proposed areas include serious incidents, near misses, and loss of control scenarios. Existing proposals already mandate reporting duties for certain regulated operators and digital service providers, with initial cyber incident notifications due within 24 hours, followed by more detailed reports within 72 hours under the planned framework.
Prioritization of AI safety measures in national security efforts
Narayan underscored Britain’s need for enhanced cyber defenses, independent technical evaluations, and domestic expertise in advanced AI. He also indicated that legislation might be required if current safeguards prove insufficient against emerging risks. The government is working on additional support mechanisms for AI verification and intends to publish further details in the coming months. These verification methods are designed to assess whether systems operate as intended and to ensure security controls remain effective even under challenging or atypical conditions.
These initiatives are part of a broader governmental review of artificial intelligence’s role and impact on national interests. The Prime Minister’s AI Taskforce is analyzing issues related to security, public services, economic growth, and workforce adaptation. Britain has also included AI on the agenda for its 2027 G20 presidency. The recent measures concentrate on addressing software vulnerabilities, evaluating advanced models, and enhancing preparedness for serious AI-related incidents across government infrastructures.